Change search
ReferencesLink to record
Permanent link

Direct link
Semantic Segmentation of Satellite Images using Deep Learning
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
2016 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

A stark increase in the amount of satellite imagery available in recent years has made the interpretation of this data a challenging problem at scale. Deriving useful insights from such images requires a rich understanding of the information present in them. This thesis explores the above problem by designing an automated framework for extracting semantic maps of roads and highways to track urban growth of cities in satellite images. Devising it as a supervised machine learning problem, a deep neural network is designed, implemented and experimentally evaluated. Publicly available datasets and frameworks are used for this purpose. The resulting pipeline includes image pre-processing algorithms that allows it to cope with input images of varying quality, resolution and channels. Additionally, we review a computational graph approach to building a neural network using the TensorFlow framework.

Place, publisher, year, edition, pages
Keyword [en]
Satellite Imagery, Deep Learning, Semantic Segmentation, Machine Learning, Urban Growth
URN: urn:nbn:se:ltu:diva-38558OAI: diva2:1013270
Educational program
Space Engineering, master's level
Available from: 2016-10-04 Created: 2016-10-03 Last updated: 2016-10-04Bibliographically approved

Open Access in DiVA

fulltext(17577 kB)97 downloads
File information
File name FULLTEXT01.pdfFile size 17577 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
Department of Computer Science, Electrical and Space Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 97 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

ReferencesLink to record
Permanent link

Direct link